Effects of the parasitic flies of the genus Philornis (Diptera:Muscidae) on birds

Transcription

1 CSIRO PUBLISHING Emu, 2006, 106, Effects of the parasitic flies of the genus Philornis (Diptera:Muscidae) on birds Rachael Y. Dudaniec A and Sonia Kleindorfer A,B A School of Biological Sciences, Flinders University, Bedford Park, Adelaide, SA 5042, Australia. B Corresponding author. sonia.kleindorfer@flinders.edu.au Abstract. Little is known about the genus Philornis (comprising ~50 species), a group of muscid flies that parasitise birds and may be highly detrimental to host nestlings. Philornis species affect at least 115 species of bird, particularly in the Neotropics. The main distribution of Philornis is in Central and South America, extending to the southern United States. Larvae of the genus Philornis reside in bird nests and may feed on either nestling faeces (coprophagous scavengers), the blood of nestlings (semi-haematophagous parasites), or on nestling tissue and fluid (subcutaneous parasites). Depending on the species of Philornis, larval development can occur in bird faeces, in nesting material or inside nestlings. Nestling mortality depends on the species of Philornis, the intensity of infestation and nestling susceptibility, which in turn depends on the nestling species, age, brood size, body condition and the anatomical site of infestation. Consequently, variable effects of Philornis parasitism are observed in relation to nestling growth, development and fledging success. The impetus for this review is the recent discovery of Philornis downsi on the Galapagos Archipelago, combined with high Philornis-induced mortality in Darwin s finches. The potential for ectoparasites such as Philornis to compromise the viability of small, isolated bird populations is highlighted by this recently documented parasite invasion. Introduction In 1997, Fessl et al. (2001) observed the presence of the introduced obligate bird ectoparasite, Philornis downsi (Diptera: Muscidae), in nests of Darwin s finches on the Galapagos Archipelago. P. downsi is characterised by freeliving non-parasitic adults whose larvae develop in bird nests as semi-haematophagous (bloodsucking) external parasites on nestlings (Couri 1985, 1999; Fessl and Tebbich 2002). The date of introduction of the parasite to the Galapagos remains speculative. Until very recently, the earliest known occurrence of P. downsi was 1981, but this has been predated by the identification of six archived specimens of P. downsi from the island of Santa Cruz from collections made by D. Q. Cavagnaro and R. C. Schuster in 1964 (California Academy of Sciences, San Francisco; Causton et al. 2006). Philornis downsi infestation is reported to cause significant fitness costs in Darwin s finches. For example, high mean number of P. downsi parasites per nestling (parasite intensity; Bush et al. 1997) was associated with higher nestling mortality (19% total brood loss, and 8% partial brood loss) in a comparison of four Darwin finch species (Fessl and Tebbich 2002), and low haemoglobin levels in the Small Ground-Finch (Geospiza fuliginosa) (Dudaniec et al. 2006). An experimental study subsequently showed impaired nestling development and reduced mass gain over a 4-day period owing to Philornis parasitism in two species of Darwin s finches as well as 62% parasite-induced mortality Royal Australasian Ornithologists Union 2006 (Fessl et al. 2006) (Table 1). Although no conclusive evidence is available, P. downsi may be implicated in the suspected local extinction of Darwin s Warbler Finch (Certhidea fusca) on Floreana Island by 2004 (Grant et al. 2005). Infestation by P. downsi is widespread on Santa Cruz Island: larvae and puparia were found in 97% of 177 nests of 12 species of birds examined in 1998 and 2000 (Fessl and Tebbich 2002). The parasite was associated with total or partial brood mortality in 27% of 85 active nests, with significant differences in parasite intensity among species (Fessl and Tebbich 2002). Since 1998, P. downsi has been found on all inhabited islands of the Galapagos (Wiedenfeld et al., in press). The potentially catastrophic effects of P. downsi on the Galapagos finches are the impetus for this review. Only by understanding the parasite and its effects can conservation measures be implemented to preserve an iconic group of birds that are widely regarded as an evolutionary treasure (Grant 1999). Little is known about the ecology and biology of Philornis flies, as most previous work has concentrated on taxonomy (Aldrich 1923; Dodge 1955, 1968; Dodge and Aitken 1968; Couri 1984, 1999; Skidmore 1985; Carvalho 1989; Couri and Carvalho 2003). Only a handful of studies have explored host Philornis biological associations (e.g. Kinsella and Winegarner 1974; Couri 1985; Fraga 1984; Teixeira 1999; Spalding et al. 2002). Given the paucity of information on P. downsi, observations from studies of other Philornis species (e.g. Oniki 1983; Arendt 1985a; Delannoy /MU /06/010013

2 14 Emu R. Y. Dudaniec and S. Kleindorfer Table 1. Characteristics of hosts (adult body length, clutch size, nest type) affected by Philornis parasitism where information on parasite intensity, mortality or other fitness impacts is available in the literature Percentage mortality is expressed in parentheses as the number of nestlings that died out of the number of observed infested nestlings Philornis species Host species Range of parasite Body length Host Host Percentage Other fitness impacts Reference intensity per (cm) of host clutch-size nest-type mortality of nestling (n = total as adult infested nestlings infested nestlings) Semi-haematophagous species P. downsi Darwin s finches Dome 27 (total or partial Nests with small broods had higher Fessl and Tebbich 2002 (four species) (n = 49) mortality) (23/85) parasite intensity and reduced fledging success Small Ground-Finch (Geospiza Dome 62 (8/13) Reduced mass gain in nestlings and Fessl et al fuliginosa) and Medium (n = 13) reduced fledging success Ground-Finch (G. fortis) Small Ground-Finch Dome 29 (17/59) Lower haemoglobin level, increased Dudaniec et al (G. fuliginosa) (n = 59) reticulocyte numbers, and reduced fledging success Rufous-capped Antshrike Cup Not recorded Mendonça and Couri (Thamnophilus ruficapillus) (n = 1 nest) 1999 Subcutaneous species P. carinatus House Wren Cavity 0 19 (42 nests) Shorter wing chords and tarsi at fledging Young 1993 (Troglodytes aedon) (n > 140) and reduced mass on day 12 P. deceptiva Pearly-eyed Thrasher Cavity 46.7 (209/448) Delayed growth and development Arendt 1985b (Margarops fuscatus) (n = 448) (body mass and tarsus length) 64 and (2/2) Loss of red blood cells, connective Uhazy and Arendt 1986 (n = 2) tissue and tissue fluids, evidence of cellular immune response P. glaucinus Chestnut-backed Antshrike 1 4 (n = 1) Hanging basket Not recorded Loss of upperwing coverts owing Mendonça and Couri (Thamnophilus palliatus) to presence of larvae 1999 P. pici and Brown Cacholote 3 21 ~ Dome 30.7 (8/26) Nores 1995 P. seguyi (Pseudoseisura lophotes) (n = 26) Firewood-gatherer 3 17 ~18 Dome 31 (9/29) Nores 1995 (Anumbius annumbi) (n = 29) Neomusca (Philornis) Great-crested Flycatcher 1 21 ~ Cavity 3.1 (1/32) Kinsella and Winegarner porteri (Myiarchus crinitus) (n = 32) 1974 Philornis sp. Puerto Rican Sharp-shinned Hawk Dome 61 (25/41) Delannoy and Cruz 1991 (unidentified) (Accipiter striatus venator) (n = 41) Mourning Dove 17 and Stick platform 0 (0/2) Glasgow and Henson (Zenaidura macroura) (n = 2) 1957 Great Kiskadee ~ Dome 0 (0/4) Physical deformation Oniki 1983 (Pitangus sulphuratus) (n = 4) Beechey Jay (Cyanocorax Cup 33 (1/3) Delayed behavioural development; Winterstein and Raitt [Cissilopha] beecheii) (n = 3) physical debilitation; reduced lengths 1983 of primary 9 and rectrix 1 Aplomado Falcon Cup Not recorded Hector 1982 (Falco femoralis) (n = 3) Masked Gnatcatcher 3 in dead ~ Cup 50 (1/2) Fraga 1984 (Polioptila dumicola) nestling (n = 2) Firewood-gatherer 5 in dead ~18 Dome 50 (2/4) Fraga 1984 (Anumbius annumbi) nestlings (n = 4) Chalk-browed Mockingbird 8 in dead Cup 33.3 (1/3) Fraga 1984 (Mimus saturninus) nestling (n = 3) Screaming Cowbird Brood (Host nest) 0 (0/3) The host species removed larvae Fraga 1984 (Molothrus rufoaxillaris) (n = 3) parasite Dome, chamber from nestlings

3 Effects of parasitic Philornis flies on birds Emu 15 and Cruz 1991; Nores 1995) are vital to increase our current understanding of this parasite. It is important to point out, however, that none of the other species of Philornis studied had semi-haematophagous larvae like P. downsi, which curtails the possibility of direct comparisons. This review provides a synthesis of the current state of knowledge regarding the general biology and fitness costs of Philornis parasitism in birds. We examine the systematics of the genus Philornis, its distribution and patterns of host choice, the biology of some Philornis species and impacts of Philornis species on nestlings in relation to parasite intensity, nestling mortality, growth and development, brood size and fledging success. Systematics of the genus Philornis The first species of what is now the muscid fly genus Philornis was described as Aricia pici from the Dominican Republic (Macquart 1854). Two additional species were described before the new genus Philornis was proposed for the fourth species P. molesta (Meinert 1890). Initially, Philornis was confused and synonymised with Protocalliphora (Diptera: Calliphoridae), another genus of birdinfesting parasitic flies, until distinctive diagnostic features were recognised and it was properly separated and placed in the family Muscidae (Bezzi 1922). Generic arrangement within the family is still in flux, but the most recent placement of the genus is in the subfamily Azeliinae, tribe Reinwardtiini (Skidmore 1985; Couri and Carvalho 2003; Carvalho et al. 2005). Philornis (with ~50 species) and the related genus Passeromyia (five species) are the only known Muscidae whose larvae are consistently parasites of birds, although this avian association seems to have arisen independently in each group (Couri and Carvalho 2003). The two genera, together with several other parasitic and non-parasitic genera, belong to a monophyletic group characterised by a single synapomorphic character: a puparium enclosed in a cocoon (Couri and Carvalho 2003). Most Philornis species were described in the 1960s (Dodge and Aitken 1968) and the 1980s (Couri 1984), and keys have only recently become available to identify most of the adults (Couri 1999). The immature stages of Philornis and their avian host relationships are known for only about half of the named species (Couri 1999; Teixeira 1999), and many previous biological studies faced taxonomic difficulties in recognising the species under study (Hector 1982; Oniki 1983; Winterstein and Raitt 1983). Dodge and Aitken (1968) pointed out that several Philornis species may occur in a variety of geographical forms or subspecies (i.e. they are polytypic). Distribution and host choice The ~50 species of Philornis currently known occur throughout the Neotropical Region, and the distribution of up to six species (including P. porteri and P. obscura) extends to the United States (Dodge 1955; Spalding et al. 2002; Couri and Carvalho 2003). Species of Philornis have been collected in Texas, Louisiana, Florida, Mexico, Guatemala, Costa Rica, Panama, Cuba, Dominican Republic, Puerto Rico, Trinidad, Venezuela, Guyana, Ecuador, Peru, Brazil, Uruguay, and Argentina (Dodge 1955, 1968; Dodge and Aitken 1968; Couri 1999; Fessl et al. 2001). Philornis species have been reported to infest at least 127 species of birds (Couri 1985; Teixeira 1999; Fessl et al. 2001; Fessl and Tebbich 2002) and they do not show marked host specificity (Couri 1991; Teixeira 1999). Known hosts for Philornis spp. are mainly Neotropical passerines (listed in Dodge 1955, 1968; Dodge and Aitken 1968; Teixeira 1999), but some infestation has been found in Falconiformes (Hector 1982; Delannoy and Cruz 1991), Galbuliformes, Cuculiformes, Galliformes, Columbiformes, Psittaciformes, Apodiformes, Piciformes and Strigiformes (Teixeira 1999). The species affecting Darwin s finches, P. downsi, is known from collections in Trinidad and Brazil (Dodge and Aitken 1968; Uhazy and Arendt 1986; Mendonça and Couri 1999) and has been reported to infest 26 species of birds in 22 mostly passerine genera, including those on the Galapagos Archipelago (Fessl et al. 2001). The biology and ecology of Philornis Little is known about the biology of Philornis species, as general information is limited to 28 (56%) of the ~50 described species (Teixeira 1999). Adult Philornis flies are non-parasitic and feed on decaying organic matter, fruit or flowers (Teixeira 1999; Fessl et al. 2001). Larval trophic relationships are documented for only 22 species. Apart from the larvae of two species that are coprophagous scavengers, the 20 remaining Philornis species for which the larval biology is known have been associated with parasitism in a diverse array of New World bird taxa. P. vespidicola is an exception as it is only known from a wasp nest (Paracharitopus frontalis (Hymenoptera: Vespidae)), but this is considered an unusual and inadequately explained observation (Dodge 1968; Teixeira 1999). Where known, the larval habits of Philornis species are divided into three groups: coprophagous, semihaematophagous and subcutaneous. Most (82%) are parasitic subcutaneous tissue and fluid feeders (e.g. P. deceptiva, P. augustifrons and P. glaucinis), which burrow into the host s integument and reside beneath the skin between the dermis and body musculature, resulting in the formation of individual cysts (Teixeira 1999; Spalding et al. 2002). These endoparasitic larvae feed on serous fluids, tissue debris and blood of the host, and each larva breathes through a small aperture it cuts in the host s integument (Skidmore 1985; Uhazy and Arendt 1986; Young 1993). Larvae of some Philornis species (9%) are free-living semi-haematophagous parasites (e.g. P. downsi and P. falsifica), which also live freely and develop in the host s nesting material. However,

4 16 Emu R. Y. Dudaniec and S. Kleindorfer the larvae periodically visit the integument of nestling hosts to feed, which they do by cutting an opening in the skin and subsequently ingesting blood and fluids of the host (Dodge and Aitken 1968; Teixeira 1999). The free-living larvae of commensal coprophagous species (9%; e.g. P. aitkeni and P. rufoscutellaris) feed and develop in accumulated organic debris (primarily faeces) at the bottom of nests of certain Neotropical birds that nest in closed cavities (Dodge 1963; Couri 1999; Teixeira 1999; Couri and Carvalho 2003). All Philornis species choose hosts with altricial or semialtricial young. Apart from coprophagous species that preferentially infest cavity nests with increased organic matter, the parasites do not select hosts with particular types of nest, having been found in cup-shaped, domed and cavity nests (Table 1). Given this lack of specific host selection it is not surprising that some hosts are affected by more than one species of Philornis that differ in parasitic strategy (Oniki 1983; Teixeira 1999). Although most muscids are oviparous, viviparity occurs commonly in the tribe Reinwardtiinae (Skidmore 1985). The Reinwardtiinae includes the species infesting Darwin s finches, P. downsi, but there is no definitive evidence regarding the reproductive habit of this species. In Darwin s finches, P. downsi larvae have been collected from the nares of young nestlings (B. Fessl, B. J. Sinclair and S. Kleindorfer, unpublished data). This suggests that P. downsi larvae develop in nestling nares before moving into the nesting material, where they have been observed later in the nesting cycle (Fessl and Tebbich 2002). Subcutaneous larvae of P. porteri have also occasionally been observed in the nares of Great-crested Flycatchers (Myiarchus crinitus) in Florida (Kinsella and Winegarner 1974), though perhaps incidentally given the endoparasitic habit of this species. Although Philornis infestations have been observed at all stages of the nestling feeding phase (Arendt 1985a; Young 1993; Nores 1995), no Philornis larvae have been observed in nests before the eggs of the hosts hatch (Arendt 1985b; Young 1993; Fessl and Tebbich 2002). Rather, larval infestation occurs within hours or a few days of nestlings hatching (Kinsella and Winegarner 1974; Spalding et al. 2002). Larval feeding and growth are completed within 4 8 days in endoparasitic species and up to 29 days in coprophagous species (Teixeira 1999). During this time larvae grow to 1 cm or more in length (this differs across Philornis species) and pass through three stadia (Fraga 1984; Arendt 1985a; Skidmore 1985; Delannoy and Cruz 1991; Spalding et al. 2002). Larvae then drop from the host to the base of the nest, where each forms a frothy cocoon (or puparium) from salivary gland secretions and in which they pupariate (Dodge 1971; Skidmore 1985), before undergoing pupation (development into adult stage). Larvae exit (in subcutaneous species) or detach from chicks (in semi-haematophagous species) quickly after nestling mortality or before fledging, and proceed to burrow into the nesting material (Teixeira 1999; Spalding et al. 2002). Such larvae may be able to pupariate before they are fully grown, based circumstantially on variations in size among pupae and adult flies (Kinsella and Winegarner 1974; Spalding et al. 2002). The duration of the pupal stage may vary from 5 to 15 days according to species and environmental conditions, though the majority of species take approximately 2 weeks to emerge as adults (Glasgow and Henson 1957; Oniki 1983; Delannoy and Cruz 1991; Spalding et al. 2002). Despite the high degree of host generalism found among parasitic Philornis, the life-cycles of Philornis species seem to be closely synchronized to the host nestling phase, as the flies require a living host with a breeding period complementary to their own relatively short lifespans (Teixeira 1999). Adult P. downsi flies were observed emerging from nests of Darwin s finches within days of hosts fledging (S. Kleindorfer and B. Fessl, unpublished data). Oniki (1983) observed that larvae of several Philornis species had detached from nestlings and started pupation in the nesting material a few days before host fledging. Philornis species parasitising multiple bird species might adjust the duration of their larval periods to match the varying nestling periods of their hosts, choose to attack only the youngest nestlings, or be able to pupate outside the nest, as some species do when they occasionally parasitise adult bird hosts (Oniki 1983). Repeated fly infestations of nests occur throughout the nestling period, and larvae of different instars have been observed simultaneously in individual nests (Oniki 1983; Arendt 1985b; Nores 1995), perhaps indicating the number of fly cohorts a nest might produce (e.g. Young 1993). Observed fluctuations in the number of parasites per nestling over the nestling period also suggest repeated fly infestations (Winterstein and Raitt 1983). Larval numbers per nestling increased with nestling age during the nestling feeding phase in Pearly-eyed Thrashers (Margarops fuscatus) (Arendt 1985b). A brood of nestling House Wrens (Troglodytes aedon) can sustain up to three or four cohorts of Philornis flies, although single cohorts were predominantly observed (Young 1993). There was no indication of larval reinfestation in nests of Darwin s finches treated with a 1% pyrethrin solution when the nestlings were 4 6 days old, which effectively reduced P. downsi intensity to less than two larvae per nest (Fessl et al. 2006). Philornis parasitism of adult birds has been reported, but it is not comparable with the level or impact of nestling infestations. Of 105 adult Pearly-eyed Thrashers, 31% were infested with P. deceptiva, but no evidence of parasiteinduced mortality was observed (Arendt 1985b). Parasitism of adult birds by Philornis species may be opportunistic and is most probably limited by host mobility (Teixeira 1999), feather protection (e.g. in nestlings, Oniki 1983) or sex (Arendt 1985b) or combinations of these factors. In Pearlyeyed Thrashers, infestation was much higher in females rearing young (46.7%) than in males (13.3%) who do not

5 Effects of parasitic Philornis flies on birds Emu 17 spend long periods in the nest. In the absence of convincing evidence to the contrary, nestling birds seem to be the primary hosts of Philornis (Teixeira 1999). Impacts of Philornis on nestlings Mortality and fledging success There have been few studies on the impact of Philornis parasitism in nestling birds. Whereas some studies have found significant reductions in nestling fitness and survival caused by Philornis larvae (Winterstein and Raitt 1983; Delannoy and Cruz 1991; Young 1993; Nores 1995; Dudaniec et al. 2006; Fessl et al. 2006), others have not (Glasgow and Henson 1957; Kinsella and Winegarner 1974; Oniki 1983). This discrepancy in findings may be partly explained by variation in host species, parasite species, and environmental conditions (Teixeira 1999). Table 1 provides an overview of some of the main variables that may affect the fitness costs of Philornis parasitism and their impact between avian hosts. The highest mortality levels (>50% of broods with total or partial mortality) among birds in relation to Philornis parasitism were found by Fraga (1984), Delannoy and Cruz (1991) and Fessl et al. (2006). Philornis downsi on the Galapagos Islands was associated with 62% brood loss in Darwin s finches (Fessl et al. 2006), whereas fledging success differed markedly between brood sizes of one (0%), two (~50%) and three or four (75 85%) chicks (Fessl and Tebbich 2002). Notably, two other parasitic Diptera were commonly found in the Darwin finch nests sampled (Sarcodexia lambdens and an unidentified endoparasitic Muscidae species), thus the results may not solely reflect the activities of P. downsi (Fessl and Tebbich 2002). The broodsize dilution effect observed by Fessl and Tebbich (2002) is a widely recognised phenomenon in avian ectoparasitism (Richner and Heeb 1995). Delannoy and Cruz (1991) found an almost 4-fold difference in mortality between Philornis-parasitised (61% brood mortality) and unparasitised Puerto Rican Sharpshinned Hawk (Accipiter striatus venator) nestlings. Fraga (1984) found 50% partial brood mortality in Philornisparasitised broods of Masked Gnatcatchers (Polioptila dumicola) and Firewood-gatherers (Anumbius annumbi), with 33% loss in Chalk-browed Mockingbirds (Mimus saturninus). Nearly half of 448 Pearly-eyed Thrasher nestlings infested with P. deceptiva died over a 4-year study period, whereas fledging success was approximately 2-fold higher in unparasitised (98%) than in parasitised nests (42 56%) (Arendt 1985b). In contrast, Nores (1995) found that fledging success in Brown Cacholotes (Pseudoseisura lophotes) and Firewoodgatherers did not differ significantly between nestlings that were simultaneously parasitised by Philornis pici and P. seguyi (69% fledged) and those that were unparasitised (75% fledged). In addition, total parasite-induced mortality was relatively low (5.5 and 5.6% in Brown Cacholotes and Firewood-gatherers respectively) owing to low parasite prevalence (i.e. percentage of hosts infested; 16%), though a third of the infested nestlings died (Table 1). Although behavioural development was delayed in nestling Purplishbacked Jays (Cyanocorax (Cissilopha) beecheii) infested with Philornis, survival to 1-year old was not affected by the level of nestling parasitism (Winterstein and Raitt 1983). The lack of evidence linking Philornis parasitism with elevated mortality across some studies has been attributed to inconclusive evidence (Oniki 1983), small sample size (Hector 1982) or the absence of any mortality (Glasgow and Henson 1957; Young 1993). Clearly, the variation in fitness costs resulting from Philornis parasitism should be examined in the context of parasite species and intensity as well as body size, clutch-size and nesting habit of the hosts (see Table 1). Parasite intensity The correlation between nestling mortality and the number of Philornis larvae infesting the brood does not show a consistent relationship, either within or between host species. The single most important factor likely to explain this discrepancy is the species of both host and parasite (see also discussion above) (Table 1). But even with constancy in these two basic elements (host and parasite species), variation in mortality under different parasite intensities may derive from other factors, such as nestling age, size, nutritional condition or anatomical site of larval infestation (Delannoy and Cruz 1991). Environmental stochasticity may generate further variation, such as in rainfall, which was found to positively influence infestation prevalence (Delannoy and Cruz 1991; Arendt 2000). Notably, all information on this subject is derived from studies of endoparasitic Philornis species, creating an imperative for further research into species with ectoparasitic and coprophagous habits. Mortality among bird hosts resulting from Philornis infestation differs according to mean parasite intensity. Four nestlings of Great Kiskadee (Pitangus sulphuratus) observed with individual loads of unidentified Philornis larvae were deformed but not dying (Oniki 1983), but other studies found that larval loads of five (Arendt 1985b), six (Delannoy and Cruz 1991) or 13 (Nores 1995) per nestling were sufficient to cause mortality in different host species, though the mass of nestling hosts might be a determining factor here. Arendt (1985a) found that when larvae infested sensitive areas around the nestling s head, less than five larvae were sufficient to cause debilitation in Pearly-eyed Thrashers. Brood parasitism may play an interesting role in reducing the negative impact of Philornis for both the avian host and brood parasite. For example, Fraga (1984) reported adult Baywinged Cowbirds (Molothrus badius) removing Philornis larvae from their nestlings and those of their coexisting brood parasite, the Screaming Cowbird (Molothrus rufoaxillaris). The high fitness benefits associated with this parental care

6 18 Emu R. Y. Dudaniec and S. Kleindorfer may explain the host specialisation of this brood parasite (Fraga 1984). Parasite prevalence may increase towards the end of the breeding season (Arendt 1985a, 1985b; Young 1993), perhaps owing to building seasonal fly populations, although not all studies have found this trend (Nores 1995; Fessl and Tebbich 2002). Such variation in prevalence as well as intensity have been proximately related to temporal variation in resource abundance (e.g. related to rainfall) (Arendt 1985b, 2000), resulting in changes in fly densities (Delannoy and Cruz 1991). Nestling growth and development Nestling Pearly-eyed Thrashers parasitised by Philornis experienced shorter and delayed growth increments in the tarsus and ninth primary pin-feather during the first week of life compared with unparasitised nestlings (Arendt 1985a). Growth of flight components, including the ulna and exposed shaft of the ninth primary were not adversely affected by parasitism, which suggests that birds maintain sufficient energy transfer to flight components under physiological stress (Arendt 1985a). However, in another study, lengths of the ninth primary and first rectrix were found to be significantly greater in uninfested than infested nestling Purplish-backed Jays (Winterstein and Raitt 1983). Young (1993) found that parasitised nestling House Wrens fledged with significantly shorter wings and showed a trend towards shorter tarsi than parasite-free nestlings. Body mass may be negatively affected by Philornis parasitism, after controlling for increases in weight owing to larval biomass (Winterstein and Raitt 1983; Arendt 1985a). In some cases, however, infested and uninfested nestlings had similar mass at fledging (Young 1993). A recent study of Darwin s finches found that nestlings in experimentally manipulated parasite-free nests (treated with 1% pyrethrin solution) had an almost 2-fold positive difference in mass across 4 days compared with untreated nests during the feeding phase (Fessl et al. 2006). Such strong effects may be explained by the recent introduction of P. downsi to the Galapagos Archipelago, as hosts may not have developed strong behavioural or immunological defence mechanisms. Furthermore, island populations may be particularly susceptible to parasitism because the ability of hosts to disperse is restricted, allowing high prevalence and impact of parasitism (Price 1980; Delannoy and Cruz 1991). Nestling vulnerability to parasitism Vulnerability and survival of nestlings parasitised by Philornis has been related to host age, with stronger negative impacts in younger nestlings (Nores 1995). Older nestlings are thought to be less vulnerable because they have greater mobility, might be partly protected by feathers, or may actively deter flies, whereas younger nestlings are naked, less active, and perhaps have softer tissue sites for larval entry (Hector 1982; Oniki 1983; Nores 1995; Teixeira 1999). Hatchlings and 1-week-old nestlings of the Brown Cacholote and Firewood-gatherer bore greater numbers of larval P. pici and P. seguyi than did older nestlings (Nores 1995), and increased intensity was correlated with increased mortality in other host Philornis studies involving the Pearly-eyed Thrasher (Arendt 1985b) and Puerto Rican Sharp-shinned Hawk (Delannoy and Cruz 1991). It should be noted that the identification of P. sequyi in Nores (1995) may be in doubt as there is a lack of information for this species (e.g. see Teixeira 1999). Site-specificity (i.e. the preferred anatomical location) of Philornis larvae on nestlings also varies with nestling age, and may influence mortality because body areas vary in sensitivity to injury (Delannoy and Cruz 1991). Again, the species of Philornis and host involved is likely a major underlying component of this variability. For example, anatomical site-specificity will depend on whether the species of adult Philornis deposits its eggs or larvae in the nesting material or directly onto nestlings, as the former is likely to be associated with parasitism of ventral surfaces (Arendt 1985a). Unfortunately, the only information found on the subject of site-specificity is limited to endoparasitic Philornis species, though larvae of semi-haematophagous P. downsi have been positively identified inside the nares of Darwin s finch nestlings (B. J Sinclair, B. Fessl and S. Kleindorfer, unpublished data), and this has not been reported in endoparasitic species. Arendt (1985b) and Uhazy and Arendt (1986) found that young Pearly-eyed Thrasher nestlings (1 9 days old) had more larvae on the head, mouth and dorsal areas of the trunk, but larvae were concentrated on the legs and ventral surfaces of older nestlings (10 19 days old). This suggests that adult flies may exploit readily accessible sites for oviposition (Uhazy and Arendt 1986). Sitespecificity may disappear later in the nestling period, when populations of both flies and hosts are most dense (Arendt 1985b). However, Nores (1995) found no significant variation in anatomical distribution of Philornis larvae during nestling development in Brown Cacholotes and Firewoodgatherers, and Oniki (1983) also found that larvae infested all ages of several hosts examined, without notable preference for sites. Conclusion A review of the available literature on Philornis flies confirms the paucity of ecological and biological information available for this genus. For example, very little is known about the adult stage in Philornis, and little is known of its reproductive biology (Arendt 1985b; Skidmore 1985; Teixeira et al. 1990; Teixeira 1999). Although a handful of studies have examined avian host Philornis parasite interactions in particular, much work remains to be done. The fitness costs of Philornis parasitism in birds may be severe, with high incidences of nestling mortality, although

7 Effects of parasitic Philornis flies on birds Emu 19 this is not always the case (Table 1). Host populations that are newly colonised by Philornis parasites, such as Darwin s finches on the Galapagos Archipelago, may experience significant initial mortality costs. These observations raise some important questions concerning the sustainability of small, isolated bird populations under threat from parasite invasion, and the potential for avian hosts to develop an adaptive response to introduced pathogens (Altizer et al. 2003). The study of P. downsi on the Galapagos Archipelago provides a useful opportunity to increase our understanding of avian parasite coevolution in the wild, particularly on species-poor islands where resistance to alien insect introductions is often low (Causton et al. 2006). In birds, ectoparasites have been linked with mediating selection on life-history trade-offs at morphological, behavioural and physiological levels (Richner 1998). The available information on host Philornis interactions reviewed here suggests that Philornis parasites contribute substantially to these associations. Acknowledgments We thank Birgit Fessl, David Wiedenfeld and Bradley Sinclair for stimulating discussion about the distribution, ecology and taxonomy of Philornis. This manuscript greatly benefited from the insightful comments made by Jeremy Robertson and two anonymous reviewers. References Aldrich, J. M. (1923). The genus Philornis a bird-infesting group of Anthomyiidae. Annals of the Entomological Society of America 16, Altizer, S., Harvell, C. D., and Friedle, E. (2003). Rapid evoutionary dynamics and disease threats to biodiversity. Trends in Ecology & Evolution 18, doi: /j.tree Arendt, W. J. (1985a). Philornis ectoparasitism of Pearly-eyed Thrashers I. Impact on growth and development of nestlings. Auk 102, Arendt, W. J. (1985b). Philornis ectoparasitism of Pearly-eyed Thrashers II. Effects on adults and reproduction. Auk 102, Arendt, W. J. (2000). Impact of nest predators, competitors, and ectoparasites on Pearly-eyed thrashers, with comments on the potential implications for Puerto Rican Parrot recovery. Ornitologia Neotropical 11, Bezzi, M. (1922). On the Dipterous genera Passeromyia and Ornithomusca, with notes and bibliography on the non-pupiparous Myiodaria parasitic on birds. Parasitology 14, Bush, A. O., Lafferty, K. D., Lotz, J. M., and Shostak, A. W. (1997). Parasitology meets ecology on its own terms: Margolis et al. revisited. Journal of Parasitology 83, Carvalho, C. J. B. (1989). Classifica Revista Brasileira de Zoología 6, Carvalho, C. J. B., Couri, M. S., Pont, A. C., Pamplona, D., and Lopes, S. M. (2005). A catalogue of the Muscidae (Diptera) of the Neotropical region. Zootaxa 860, Causton, C. E., Peck, S. B., Sinclair, B. J., Roque-Albelo, L., Hodgson, C. J., and Landry, B. (2006). Alien insects: threats and implications for conservation of Galapágos Islands. Annals of the Entomological Society of America 99, Couri, M. S. (1984). Notes and descriptions of Philornis flies (Diptera, Muscidae, Cyrtoneurininae). Revista Brasileira de Entomologia 28, Couri, M. S. (1985). Considerações sobre as relações ecológicas das larvas de Philornis Meinert, 1890 (Diptera, Muscidae) com aves. Revista Brasileira de Entomologia 29, Couri, M. S. (1991). Philornis carinatus Dodge, 1968 (Diptera, Muscidae) data on morphology, biology and taxonomy. Revista Brasileira de Entomologia 35, Couri, M. S. (1999). Myiasis caused by obligatory parasites. Ia. Philornis Meinert (Muscidae). In Myiasis in Man and Animals in the Neotropical Region; Bibliographic Database. (Eds J. H. Guimarães and N. Papavero.) pp (Editora Plêiade/FAPESP: São Paulo.) Couri, M. S., and Carvalho, C. J. B. (2003). Systematic relations among Philornis Meinert, Passeromyia Rodhain and Villeneuve and allied genera (Diptera, Muscidae). Brazilian Journal of Biology 63, doi: /s Delannoy, C. A., and Cruz, A. (1991). Philornis parasitism and nestling survival of the Puerto Rican sharp-shinned hawk. In Bird Parasite Interactions: Ecology, Evolution, and Behaviour. (Eds J. E. Loye and M. Zuk.) pp (Oxford University Press: Oxford, UK.) Dodge, H. R. (1955). New muscid flies from Florida and the West Indies (Diptera:Muscidae). Florida Entomologist 38, Dodge, H. R. (1963). A new Philornis with coprophagous larvae, and some related species (Diptera: Muscidae). Journal of the Kansas Entomological Society 36, Dodge, H. R. (1968). Some new and little known species of Philornis (Diptera: Muscidae). Journal of the Kansas Entomological Society 41, Dodge, H. R. (1971). Revisional studies of flies of the genus Philornis Meinert (Diptera, Muscidae). Studia Entomologica 14, Dodge, H. R., and Aitken, T. H. G. (1968). Philornis flies from Trinidad (Diptera: Muscidae). Journal of the Kansas Entomological Society 41, Dudaniec, R. Y., Kleindorfer, S., and Fessl, B. (2006). Haemoglobin level and nestling survival in Darwin s Small Ground Finch (Geospiza fuliginosa): effects of the introduced ectoparasite Philornis downsi. Austral Ecology 31, ). Fessl, B., and Tebbich, S. (2002). Philornis downsi a recently discovered parasite on the Galapagos Archipelago a threat for Darwin s finches? Ibis 144, doi: /j x x Fessl, B., Couri, M. S., and Tebbich, S. (2001). Philornis downsi Dodge and Aitken, new to the Galapagos Islands (Diptera, Muscidae). Studia Dipterologica 8, Fessl, B., Kleindorfer, S., and Tebbich, S. (2006). An experimental study of the fitness costs of Philornis downsi in Darwin s ground finches. Biological Conservation 127, ). Fraga, R. M. (1984). Bay-winged Cowbirds (Molothrus badius) remove ectoparasites from their brood parasites, the Screaming Cowbird (M. rufoaxillaris). Biotropica 16, Glasgow, L. L., and Henson, R. (1957). Mourning Dove nestlings infested with larvae of Philornis. Wilson Bulletin 69, Grant, P. R. (1999). Ecology and Evolution of Darwin s Finches. (Princeton University Press: Princeton, NJ.) Grant, P. R., Grant, B. R., Petren, K., and Keller, L. F. (2005). Extinction behind our backs: the possible fate of one of the Darwin s finch species on Isla Floreana, Galapagos. Biological Conservation 122, doi: /j.biocon Hector, D. P. (1982). Botfly (Diptera, Muscidae) parasitism of nestling Aplomado Falcons. Condor 84, Kinsella, J. M., and Winegarner, C. E. (1974). Notes on the life history of Neomusca porteri (Dodge), parasitic on nestlings of the Great Crested Flycatcher in Florida. Journal of Medical Entomology 11, 633.

8 20 Emu R. Y. Dudaniec and S. Kleindorfer Macquart, J. (1854). Notice sur une nouvelle espece d Aricia. Annales de la Société Entomologique de France (Ser. 3, 1853) 1, Meinert, F. (1890). Philornis molesta en paa fugle synltende tachinarie. Videnskabelige Meddelelser fra den Naturhistoriske Forening i Kjøbenhavn (Ser. 5, 1889) 1, Mendonça, E. de C., and Couri, M. S. (1999). New associations between Philornis Meinert (Diptera, Muscidae) and Thamnophilidae (Aves, Passeriformes). Revista Brasileira de Zoologia 16, Nores, A. I. (1995). Botfly ectoparasitism of the Brown Cacholote and the Firewood-gatherer. Wilson Bulletin 107, Oniki, Y. (1983). Notes on fly (Muscidae) parasitism of nestlings of South American birds. Le Gerfaut 73, Price, P. W. (1980). Evolutionary Biology of Parasites. (Princeton University Press: Princeton, NJ.) Richner, H. (1998). Host-parasite interactions and life-history evolution. Zoology 101, Richner, H., and Heeb, P. (1995). Are clutch size and brood size patterns in birds shaped by ectoparasites? Oikos 73, Skidmore, P. (1985). The Biology of the Muscidae of the World. (Dr W. Junk Publishers: Dordrecht, The Netherlands.) Spalding, M. G., Mertins, J. W., Walsh, P. B., and Morin, K. C. (2002). Burrowing fly larvae (Philornis porteri) associated with mortality of eastern bluebirds in Florida. Journal of Wildlife Diseases 38, Teixeira, D. M. (1999). Myiasis caused by obligatory parasites. Ib. General observations on the biology of species of the genus Philornis Meinert, 1890 (Diptera, Muscidae). In Myiasis in Man and Animals in the Neotropical Region; Bibliographic Database. (Eds J. H. Guimarães and N. Papavero) pp (Editora Plêiade/FAPESP: São Paulo.) Teixeira, D. M., Couri, M. S., and Luigi, G. (1990). Notas sobre a biología de Philornis rufoscutellaris Couri, 1983 (Diptera, Muscidae) es sua associação com ninhos de aves. Revista Brasileira de Entomologia 34, Uhazy, L. S., and Arendt, W. J. (1986). Pathogenesis associated with philornid myiasis (Diptera: Muscidae) on nestling pearly-eyed thrashers (Aves: Mimidae) in the Luquillo Rain Forest, Puerto Rico. Journal of Wildlife Diseases 22, Wiedenfeld, D A., Jimenez, U. G., Fessl, B., and Kleindorfer, S. (in press). Distribution of the introduced parasite fly Philornis downsi (Diptera, Muscidae) in the Galápagos Islands. Pacific Conservation Biology. Winterstein, S. R., and Raitt, R. J. (1983). Nestling growth and development and the breeding biology of the Beechey Jay. Wilson Bulletin 95, Young, B. E. (1993). Effects of the parasitic botfly Philornis carinatus on nestling house wrens, Troglodytes aedon, in Costa Rica. Oecologia 93, doi: /bf Manuscript received 13 August 2004, accepted 1 December